Hi, sorry if this sounds like a silly question; I am new in this area. I am trying to follow the example below to use a pre-trained model. The script downloads the model and stores it on my local drive (in the script directory) and everything goes well for the first time. However, an effort to run the script for the second time will fail, seemingly due to the presence of the downloaded model in the same directory. This only fixes if I delete the model (around 500 MB) from the directory before executing the script for the second time. I was wondering if there is any way I can load the model from my local drive without having to download it each time?! Also, is there any way I can use the model without having to download it, e.g., through calling a web service? I appreciate any help on this.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Download models for local loading | 11 | 94404 | March 18, 2024 | |
How Can I use cashed models from HuggingFace? | 0 | 228 | November 9, 2023 | |
"Can't load tokenizer for <model>" after download | 3 | 2712 | December 13, 2023 | |
Reloading a saved fine-tuned model trained using the Trainer Object from Huggingface does not yield correct predictions | 0 | 731 | March 15, 2022 | |
Load the model fails for sentence-transformers/sentence-t5-xl | 16 | 408 | March 31, 2025 |